The objectives of this prospective investigation are to use latent growth mixture modeling to determine the longitudinal trajectories of sexual violence (SV) behaviors during college and to use the Theory of Triadic Influence to guide the assessment of intrapersonal, social/situational, and cultural/environmental influences on the different trajectories. The rationale for the research is that it is important to identify distinct trajectories of SV behaviors during emerging adulthood when these behaviors are often initiated, and for some, escalate. More research is needed that focuses on perpetration and is longitudinal. Our team is particularly poised to undertake this investigation. This research would extend a currently-funded NIH study on SV among male college students during their first two years in college for two additional years, as well as enhance our efforts to involve students in health-related research. The specific aims are (1) to identify classes of growth trajectories of males' SV behavior during college and (2) to determine the relative importance of intrapersonal, social/situational, and cultural/environmental correlates in predicting the different SV trajectories. Findings will indicate if there are many different men who on rare occasions engage in SV behaviors or if a majority of unwanted sexual encounters are perpetrated by a small subset of males. It is hypothesized that at least three latent classes of SV exist - a high SV class characterized by consistent SV behaviors, a low SV class characterized by a consistent lack of SV behaviors, and a mixed SV class characterized by intermittent SV. It also is hypothesized that the risk factors differentiating the low SV group from the intermittent SV group will differ from those differentiating the low SV group from the high SV group. Specifically, we expect that intrapersonal factors will be associated with membership in the high SV latent class group, and that social and environmental factors will be associated with membership in the mixed SV latent class group. This study is innovative in that we will identify a subset of males who follow a trajectory of high SV behaviors and the risk factors that are prevalent among this class of individuals. The findings will advance the field by increasing understanding of how SV behavior changes over time, what predicts trajectory class membership, and ultimately, to the design of preventive interventions for SV. PUBLIC HEALTH RELEVANCE: Sexual violence (SV) is a significant public health problem. Although prevalence and risk factor studies have yielded important information about SV, there is limited longitudinal data on SV perpetration, and knowledge gaps exist regarding the pattern and predictors of different trajectories of SV behaviors as youth transition into adulthood. In order to intervene effectively to prevent SV, research is first needed to determine what factors predict the onset, maintenance, escalation, and desistance of SV behaviors during a key developmental period.